A transiting giant planet with a temperature between 250K and 430K

LETTERS

A transiting giant planet with a temperature between

250 K and 430 K

H. J. Deeg

, C. Moutou

, A. Erikson

, Sz. Csizmadia

, B. Tingley

, P. Barge

, H. Bruntt

, M. Havel

, S. Aigrain

,

J. M. Almenara

, R. Alonso

, M. Auvergne

, A. Baglin

, M. Barbieri

, W. Benz

, A. S. Bonomo

, P. Borde

´

,

F. Bouchy

, J. Cabrera

, L. Carone

, S. Carpano

, D. Ciardi

, M. Deleuil

, R. Dvorak

, S. Ferraz-Mello

,

M. Fridlund

, D. Gandolfi

, J.-C. Gazzano

, M. Gillon

, P. Gondoin

, E. Guenther

, T. Guillot

, R. den Hartog

,

A. Hatzes

, M. Hidas

, G. He

´brard

, L. Jorda

, P. Kabath

{, H. Lammer

, A. Le

´ger

, T. Lister

, A. Llebaria

,

C. Lovis

, M. Mayor

, T. Mazeh

, M. Ollivier

, M. Pa

¨tzold

, F. Pepe

, F. Pont

, D. Queloz

, M. Rabus

{,

H. Rauer

, D. Rouan

, B. Samuel

, J. Schneider

, A. Shporer

, B. Stecklum

, R. Street

, S. Udry

, J. Weingrill

& G. Wuchterl

Of the over 400 known1_{exoplanets, there are about 70 planets that}

transit their central star, a situation that permits the derivation of their basic parameters and facilitates investigations of their atmo-spheres. Some short-period planets2_{, including the first terrestrial}

exoplanet3,4_{(CoRoT-7b), have been discovered using a space mission}5

designed to find smaller and more distant planets than can be seen from the ground. Here we report transit observations of CoRoT-9b, which orbits with a period of 95.274 days on a low eccentricity of 0.11 6 0.04 around a solar-like star. Its periastron distance of 0.36 astronomical units is by far the largest of all transiting planets, yield-ing a ‘temperate’ photospheric temperature estimated to be between 250 and 430 K. Unlike previously known transiting planets, the present size of CoRoT-9b should not have been affected by tidal heat dissipation processes. Indeed, the planet is found to be well described by standard evolution models6_{with an inferred interior composition}

consistent with that of Jupiter and Saturn.

CoRoT observed the CoRoT-9 host star for 145 days starting on 15 April 2008, in the target window E1_0651 of the survey-field LRc02 in the constellation Serpens Cauda. Two transit events (Fig. 1) were detected in CoRoT’s three-colour photometry, the first one on 16 May 2008. A model fit to the first transit event indicated a Jupiter-sized planet on an equatorial transit (for values, see Table 1). Alerted by this event, two spectra of CoRoT-9 were obtained with the SOPHIE spectrograph7_{on 5 and 25 August 2008, showing a velocity difference}

of 35 6 21 m s21between both epochs, which is compatible with a giant planet.

Observations of a transit on 1 June 2009 with the Wise Observatory 1 m telescope (Israel) and the IAC 80 cm telescope (Tenerife) confirmed

that the event originated on the target star itself. A false alarm from possible nearby eclipsing binary stars, the light of which might have spread into CoRoT’s large (1699 3 2199) aperture mask, could therefore be excluded8_{. Radial velocity observations (Fig. 2 and}

Supplemen-tary Table 1) with the HARPS spectrograph9 _{spanning from 21}

September 2008 to 20 September 2009 verified a Jupiter-mass planet on a moderately eccentric orbit.

To assign reliable absolute values to the size and mass of the CoRoT-9b planet, a precise characterization of the planet host star is required. Spectra taken at the Thu¨ringer Landessternwarte and with the UVES spectrograph on the Very Large Telescope Unit 2 (ESO) indicated a G3V star of nearly solar metallicity. Its evolutionary age is not strongly constrained, ranging from 0.15 to 8 Gyr. We favour the higher end of the age range, given the quiescent light curve and the absence of chromospheric activity in the spectra. Its rotational velocity of v sin irot,3.5 km s21implies (for a rotational inclination irot< 90u) a slow rotation period of .14 days. Depletion of Li has been noted as a feature of planet-hosting stars10_{and indeed, its doublet at 6,707 A˚}

was not found in our spectra.

The combination of the light curve analysis, the radial velocity data and the determined stellar parameters allows the derivation of absolute values for the planet’s mass, radius and density (Table 1). The planet has a radius quite similar to Jupiter, with Rp5(1.05 6 0.04)RJupiter, but a lower mass of Mp5(0.84 6 0.07)MJupiter, leading to a density of 0.90 6 0.13 g cm23, or about 68% that of Jupiter.

Tidal heating is expected to play a negligible role in the planet’s evolution. The ratio of the rate of energy from tidal circularization (assuming an initial eccentricity similar to that of Jupiter) to that

1_{Instituto de Astrofı´sica de Canarias, C. Via Lactea S/N, E-38205 La Laguna, Tenerife, Spain.}2_{Department de Astrofı´sica, Universidad de La Laguna, E-38200 La Laguna, Tenerife,}

Spain.3_{Laboratoire d’Astrophysique de Marseille, CNRS and Universite}

´ de Provence, 38 rue Fre´de´ric Joliot-Curie, F-13388 Marseille cedex 13, France.4_{Institute of Planetary Research,}

German Aerospace Center, Rutherfordstrasse 2, D-12489 Berlin, Germany.5_{LESIA, Observatoire de Paris, CNRS, Place J. Janssen, 92195 Meudon cedex, France.}6_Universite

´ de Nice-Sophia Antipolis, Observatoire de la Coˆte d’Azur, CNRS UMR 6202, 06304 Nice cedex 4, France.7_{School of Physics, University of Exeter, Exeter, EX4 4QL, UK.}8_{Oxford Astrophysics,}

University of Oxford, Keble Road, Oxford OX1 3RH, UK.9_{Observatoire de l’Universite}

´ de Gene`ve, 51 chemin des Maillettes, CH 1290 Sauverny, Switzerland.10_{Dipartimento di}

Astronomia, Universita` di Padova, 35122 Padova, Italy.11_{Universita¨t Bern, Physics Institute, Sidlerstrasse 5, CH 3012 Bern, Switzerland.}12_{Institut d’Astrophysique Spatiale, Universite´}

Paris XI, F-91405 Orsay, France.13_{IAP, 98bis boulevard Arago, F-75014 Paris, France.}14_{Observatoire de Haute-Provence, CNRS/OAMP, 04870 St Michel l’Observatoire, France.} 15_{LUTH, Observatoire de Paris, CNRS and Universite´ Paris Diderot, 5 place Jules Janssen, F-92195 Meudon, France.}16_{Rheinisches Institut fu¨r Umweltforschung an der Universita}

¨t zu Ko¨ln, Aachener Strasse 209, D-50931, Ko¨ln, Germany.17_{Research and Scientific Support Department, ESTEC/ESA, PO Box 299, 2200 AG Noordwijk, The Netherlands.}18_NASA

Exoplanet Science Institute/Caltech, South Wilson Avenue, Mail Code 100-22, Pasadena, California 91125, USA.19University of Vienna, Institute of Astronomy, Tu¨rkenschanzstrasse 17, A-1180 Vienna, Austria.20Institute of Astronomy, Geophysics and Atmospheric Sciences, Universidade de Sa˜o Paulo, Brasil.21

Thu¨ringer Landessternwarte, Sternwarte 5, D-07778 Tautenburg, Germany.22

University of Lie`ge, Alle´e du 6 aouˆt 17, Sart Tilman, Lie`ge 1, Belgium.23

Las Cumbres Observatory Global Telescope Network, Inc., 6740 Cortona Drive, Suite 102, Santa Barbara, California 93117, USA.24_{Sydney Institute for Astronomy, School of Physics, The University of Sydney, New South Wales 2006, Australia.}25_{Space Research}

Institute, Austrian Academy of Science, Schmiedlstrasse 6, A-8042 Graz, Austria.26_{School of Physics and Astronomy, Raymond and Beverly Sackler Faculty of Exact Sciences, Tel}

Aviv University, Tel Aviv 69978, Israel.27

Center for Astronomy and Astrophysics, TU Berlin, Hardenbergstrasse 36, 10623 Berlin, Germany. {Present addresses: European Southern Observatory Chile, Alonso de Co´rdova 3107, Vitacura, Casilla 19001, Santiago de Chile, Chile (P.K.); Departamento de Astronomı´a y Astrofı´sica, Pontificia Universidad Cato´lica de Chile, Casilla 306, Santiago 22, Chile (M.R.).

Vol 464|18 March 2010|doi:10.1038/nature08856

384

Macmillan Publishers Limited. All rights reserved

from insolation11 _{is 30 times lower for CoRoT-9b than for}

HD80606b, and at least 1,000 times lower relative to all other transi-ting planets. Similarly, we estimate12 _{a mass loss rate of about}

8 3 106g s21, which is the lowest one among the known transiting gas giants; escape processes have therefore not affected the planet significantly since its origin. That CoRoT-9b has a much larger periastron distance than any other known transiting exoplanet also strongly constrains its composition, independently of hypotheses11,13

on possible missing energy sources and associated radius inflation: An evolutionary model of CoRoT-9b (Fig. 3) shows a good match to the observed radius between 0 and 20 Earth masses of heavy elements, comparable to the composition of giant planets in our Solar System6_.

Effective temperatures of CoRoT-9b have been calculated using a black-body approximation for the host-star emission and a uniform redistribution of absorbed heat energy between the planet’s day and night side. Following the planet classification based on temperature regimes in ref. 14, there are two possible outcomes for CoRoT-9b: It may be among class III planets with temperatures .350 K, which have clear atmospheres without cloud cover and low Bond albedos of 0.09–0.12. With such an albedo, CoRoT-9b’s temperatures would range from 380 K at apoastron to 430 K at periastron. CoRoT-9b may also be among Class II planets, with much lower temperatures (250–290 K) owing to a significantly higher Bond albedo of ,0.8 from the condensation of H2O in the upper troposphere. The strong inverse dependency of the albedo on temperatures14_{near 350 K will make it}

possible to lock the planet into either class II or class III. Transits occur not far from periastron, so temperatures during transit would be only slightly (,30 K) lower than at periastron. Day/night temperature

variations are expected to be very small for two reasons. First, from its low tidal dissipation we estimate a timescale for rotational slow-down of the order of 100 Gyr; so its rotation should still be close to the unknown, but probably rapid, primordial rate. Second, the planet’s radiative timescale at the photosphere, which is inversely proportional to the cube of the photospheric temperature15_{, should be around 50}

times longer than for a standard hot-giant planet with a 1,500 K irra-diation temperature.

Figure 4 shows a diagram of the eccentricity versus the period for all known planets and for all transiting planets. As can be seen, CoRoT-9b is the first transiting planet among those with longer periods that does not represent a case of extreme eccentricity with associated extreme temperature changes (for example, HD80806b’s temperature is estimated to rise from ,800 K to ,1,500 K over a six-hour period near periastron16_{). On the contrary, it is the first transiting planet}

whose general properties coincide with the largest-known population of planets, those of longer periods and low-to-moderate eccentricities, but which were previously known only from radial velocity surveys.

1.005 a b 1.000 Normalized flux Residuals 0.995 0.990 0.985 0.980 –0.004 –0.002 0.000 Orbital phase 0.002 0.004 0.004 –0.004 0.000

Figure 1|Light curve and model fit of the CoRoT-9b transit. a, Phase-folded light curve from CoRoT around the first observed transit of CoRoT-9b, with the phase set to zero at mid-transit. The data from the three colour-bands of CoRoT have been summed to a ‘white’ band and the points have been binned from an original cadence of 32 s to a duration of 96 s; the error bars indicate the dispersion of the points within the bins. The second transit observed by CoRoT is not shown because it was only partially detected; its first 2 h were lost in an instrument interruption. The solid line represents the model fit from which the primary fit-parameters in Table 1 have been derived. They indicate a transit close to the equator, and a stellar limb darkening that is in good agreement with predictions for the CoRoT pass

band21_{. This fit was based on the white-light data of the first transit event and}

included (1) the removal of a 2.5% contaminating fraction of light not originating from CoRoT-9, (2) the generation of a model for the planetary

transit22_{assuming a linear relation for the stellar limb-darkening, and (3) the}

fitting of the model using a heuristic optimization algorithm23_{. The}

contaminating fraction of 2.5% was determined from pre-launch imagery24_,

taken with the INT/WFC on La Palma, after convolving these images with CoRoT’s point-spread function, and measuring the contribution of nearby

stars within CoRoT’s aperture mask.b, Residuals after subtraction of the

model fit.

Table 1|Star and planet parameters of the CoRoT-9/9b system

Parameter Value ID (CoRoT-Window-Id) LRc02_E1_0651 ID (CoRoT/Exo-Dat) 0105891283 ID (GSC 2.3) N1RO059308 Position (J2000) 18:43:08.81 1 06:12:15.19 Magnitudes (ref. 17) U, B, V, r9, i9 14.74, 14.55, 13.69, 13.33, 12.86 Results from light-curve analysis

Planet period, P 95.2738 6 0.0014 days Transit epoch, T0 HJD 24 54603.3447 6 0.0001

Transit duration, T14 8.08 6 0.10 h

Relative transit depth,DF/F0 0.0155 6 0.0005

Radius ratio, Rp/Rs 0.115 6 0.001

Impact parameter, b 0:01z0:06{0:01

Limb-darkening coefficient, u 0.57 6 0.06 Scaled semimajor axis*, a/Rs 936 3 Orbital inclination*, i 89:99z0:01

{0:04

0

Stellar density*, rs 1.68 6 0.20 g cm23

Ms1/3/Rs* 1.06 6 0.04 (solar units) Results from radial velocity observations

Radial velocity semi-amplitude, K 386 3 m s21 Orbital eccentricity, e 0.11 6 0.04 Argument of periastron, v 37z9

{37u

Systemic radial velocity 19.791 6 0.002 km s21

Mass function, f (5.3 6 1.3) 3 10210_M sun

Results from spectral typing of CoRoT-9 Stellar spectral type G3V

Stellar surface gravity{, log g 4.54 6 0.09 c.g.s. Stellar effective temperature{, Teff 56256 80 K Metallicity{, [M/H] 20.01 6 0.06 Stellar mass, Mstar (0.99 6 0.04)MSun

Stellar radius, Rstar (0.94 6 0.04)RSun

Stellar rotational velocity, vsinirot #3.5 km s21

Stellar rotational period, Prot $14 days

Stellar distance 460pc Absolute physical parameters from combined analysis Planet mass, Mp (0.84 6 0.07)MJupiter

Planet radius, Rp (1.05 6 0.04)RJupiter

Planet density, rp 0.90 6 0.13 g cm23 Planet orbit semi-major axis, a 0.407 6 0.005AU

Planet–star distance at transit, at 0:377z0:025{0:015AU

* These values take into account the eccentricity as found by the radial-velocity observations. { These values are based on high-resolution (R < 65,000) spectra taken on 21 and 22 September 2008 with the ESO VLT/UVES in Dic1 mode (390 1 580), a slit width of 0.7 arcsec and analysis with the VWA software18,19_.

The planet’s quoted period was determined from the first transit observed by CoRoT and from ground-based photometry of a transit on 5 September 2009, taken with the Euler 1.2 m telescope from La Silla, Chile, and the 2 m LCOGT Faulkes Telescope North at Haleakala Observatory. For the transit duration, the time from first to fourth contact is given. The mass function indicates the ratio e 5 (Mpsini)3/(Mp1 Ms)2and is given by f 5 (1 2 e2)3/2K3P/(2pc), where c is the gravitational constant. The stellar spectral type is based on data taken with the spectrograph (R< 2,100) at the Thu¨ringer Landessternwarte (Tautenburg, Germany) on 27 July 2008. The value for the stellar surface gravity is determined directly from spectroscopy. An alternative, logg 5 4.49 6 0.04, can be derived from the stellar density given by the light-curve analysis and a weakly influencing component of Mstar1/3. The stellar mass is derived from StarEvol20_{evolutionary tracks. The stellar radius is derived from the quoted stellar mass and the}

stellar density from the transit-fit. RJupiter5 71492 km; MJupiter5 1.8986 3 1030g.AU, astronomical units.

NATURE|Vol 464|18 March 2010 LETTERS

385

Macmillan Publishers Limited. All rights reserved

Our results on CoRoT-9b show that these planets may be expected to be rather similar to the giants of our Solar System.

CoRoT-9b’s period is about ten times longer than of any other planet discovered through the transit method, which demonstrates the method’s potential to find longer-periodic planets. Further observations of such planets will, however, present new challenges. CoRoT-9b’s distance from its host-star of ,0.9 milliarcseconds is too close for imaging, and regardless of its actual Bond albedo, its relative secondary eclipse depths will be undetectable, being at the parts per million level in both the visible and the near-infrared regime. The most promising aspect of CoRoT-9b is that it will allow for spectro-scopy during primary transits, which may lead to the detection of species at moderate temperatures in the visible and infrared spectra, such as CO2at 1.25 mm, CH4at 0.95 mm, or H2O at 6 mm.

Received 30 November 2009; accepted 19 January 2010.

1. Schneider, J. The Extrasolar Planets EncyclopediaÆhttp://exoplanet.eu/index.phpæ (1999–2010).

2. Dvorak, R. et al. CoRoT’s first seven planets: An overview. In Proc. ‘‘New Technologies for Probing the Diversity of Brown Dwarfs and Exoplanets’’ (EDP publications, 2009); preprint atÆhttp://arxiv.org/abs/0912.4655æ.

3. Le´ger, A. et al. Transiting exoplanets from the CoRoT space mission. VIII. CoRoT-7b: the first super-Earth with measured radius. Astron. Astrophys.506, 287–302 (2009).

4. Queloz, D. et al. The CoRoT-7 planetary system: two orbiting super-Earths. Astron. Astrophys.506, 303–319 (2009).

5. Baglin, A. et al. CoRoT: Description of the Mission and Early Results. IAU Symp. 253, 71–81 (2009).

6. Guillot, T. The interiors of giant planets: models and outstanding questions. Annu. Rev. Earth Planet. Sci.33, 493–530 (2005).

7. Perruchot, S. et al. The SOPHIE spectrograph: design and technical key-points for high throughput and high stability. Proc. SPIE7014, 70140J (2008).

8. Deeg, H. J. et al. Ground-based photometry of space-based transit detections: photometricfollow-upof theCoRoTmission.Astron.Astrophys.506,343–352 (2009). 9. Mayor, M. et al. Setting new standards with HARPS. ESO Mess.114, 20–24

(2003).

10. Israelian, G. et al. Enhanced lithium depletion in Sun-like stars with orbiting planets. Nature462, 189–191 (2009).

11. Ibgui, L., Burrows, A. & Spiegel, D. Tidal heating models for the radii of the inflated transiting giant planets WASP-4b, WASP-6b, WASP-12b, and TrES-4. Astrophys. J. (submitted); preprint atÆhttp://arxiv.org/abs/0910.4394æ (2009). 12. Lammer, H. et al. Determining the mass loss limit for close-in exoplanets: what

can we learn from transit observations? Astron. Astrophys.506, 399–410 (2009). 13. Miller, N., Fortney, J. J. & Jackson, B. Inflating and deflating hot Jupiters: coupled

tidal and thermal evolution of known transiting planets. Astrophys. J.702, 1413–1427 (2009).

14. Sudarsky, D., Burrows, A. & Pinto, P. Albedo and reflection spectra of extrasolar giant planets. Astrophys. J.535, 885–903 (2000).

15. Showman, A. P. & Guillot, T. Atmospheric circulation and tides of ‘‘51 Pegasus b-like’’ planets. Astron. Astrophys.385, 166–180 (2002).

16. Laughlin, G. et al. Rapid heating of the atmosphere of an extrasolar planet. Nature 457, 562–564 (2009).

17. Exodat Information System.Æhttp://lamwws.oamp.fr/exodat/æ. 18. Bruntt, H. et al. Abundance analysis of targets for the COROT/MONS

asteroseismology missions. II. Abundance analysis of the COROT main targets. Astron. Astrophys.425, 683–695 (2004).

19. Bruntt, H., De Cat, P. & Aerts, C. A spectroscopic study of southern (candidate)c Doradus stars. II. Detailed abundance analysis and fundamental parameters. Astron. Astrophys.478, 487–496 (2008).

20. Siess, L. Evolution of massive AGB stars. I. Carbon burning phase. Astron. Astrophys.448, 717–729 (2006).

Figure 3|Evolutionary model for a CoRoT-9b like planet. Constraints on

the transit radius of CoRoT-9b for a given age, based on stellar25_and

planetary evolutionary tracks26_{. Its observed radius is (1.05 6 0.05)R}

Jupiter

(Table 1). Red, blue and green dots correspond to models matching the

rstar–Teffuncertainty ellipse within 1s, 2s and 3s, respectively. Planetary

evolution models for a planet with a solar-composition envelope and a central dense core of 0, 20, 40 and 60 Earth masses are shown as black, blue, green and light blue lines, respectively. These models assume that the planet is made of a solar-composition envelope over a dense icy/rocky core of variable mass. Their results depend only weakly on the assumed opacities, and uncertainties due to the atmospheric temperature, planetary mass and any tidal dissipation rate are negligible.

0 0.5 0 50 500 5,000 0.2 0.4 Eccentricity, e Period, P (day) 0.6 HD 17156b HD 80606b CoRoT-9b 0.8 1

Figure 4|The orbital parameters of CoRoT-9b among extrasolar planets.

The eccentricity and period of all 339 exoplanets for which both values are known as of 1 November 2009. Solid dots are the 58 transiting planets among them—most of them have short periods of less than about 5 days and zero or low eccentricities. Only two further transiting planets have orbital periods

longer than ten days; these are HD1715627_{with 21.2 days and HD80806}28–30_with

111.4 days. However, both of them also have the highest eccentricities among planets of similar periods. Open dots represent the remaining exoplanets, known only from radial velocity observations. Data are sourced from ref. 1. –0.4 –60 –40 –20 0 20 40 60 –0.2 0 Orbital phase Radial velocity (m s –1) 0.2 0.4

Figure 2|CoRoT-9 radial velocities. Radial velocity data from the HARPS spectrograph after subtraction of the systemic velocity, versus the orbital phase, set to zero at periapsis. The error bars on individual points are 1-sigma measurement errors, derived from the fit of a cross correlation function to the observed spectra. The solid line is the best-fit elliptical orbital

solution; the root-mean-square of its residuals is 6.3 m s21_{. This solution}

was also constrained by the known ephemeris of the planetary transits. An F-test comparing a best-fit circular orbit to the adopted elliptic orbit gives 95% confidence for the presence of an ellipticity that is significantly different from zero.

LETTERS NATURE|Vol 464|18 March 2010

386

Macmillan Publishers Limited. All rights reserved

21. Sing, D. K. Stellar limb-darkening coefficients for CoRoT and Kepler. Astron. Astrophys.510, 21–27 (2010).

22. Mandel, K. & Agol, E. Analytic light curves for planetary transit searches. Astrophys. J.580, L171–L175 (2002).

23. Geem, Z. W., Kim, J. H. & Loganathan, G. V. A new heuristic optimization algorithm: Harmony Search. Simulation76, 60–68 (2001).

24. Deleuil, M. et al. Exo-Dat: an information system in support of the CoRoT/ exoplanet science. Astron. J.138, 649–663 (2009).

25. Morel, P. & Lebreton, Y. CESAM: a free code for stellar evolution calculations. Astrophys. Space Sci.316, 61–73 (2008).

26. Guillot, T. The composition of transiting giant extrasolar planets. Phys. Scr.130, 014023 (2008).

27. Barbieri, M. et al. HD 17156b: a transiting planet with a 21.2 day period and an eccentric orbit. Astron. Astrophys.476, L13–L16 (2007).

28. Fossey, S. J., Waldmann, I. P. & Kipping, D. M. Detection of a transit by the planetary companion of HD 80606. Mon. Not. R. Astron. Soc.396, L16–L20 (2009).

29. Moutou, C. et al. Photometric and spectroscopic detection of the primary transit of the 111-day-period planet HD 80606 b. Astron. Astrophys.498, L5–L8 (2009). 30. Garcia-Melendo, E. & McCullough, P. R. Photometric detection of a transit of HD

80606b. Astrophys. J.698, 558–561 (2009).

Supplementary Information is linked to the online version of the paper at www.nature.com/nature.

Acknowledgements The CoRoT space mission has been developed and is operated by CNES, with the contributions of Austria, Belgium, Brazil, ESA, Germany

and Spain. CoRoT data are available to the public from the CoRoT archive: http:// idoc-corot.ias.u-psud.fr. The team at IAC acknowledges support by grant ESP2007-65480-C02-02 of the Spanish Ministerio de Ciencia e Innovacio´n. The German CoRoT Team (TLS and Univ. Cologne) acknowledges DLR grants 50OW0204, 50OW0603 and 50QP07011. Observations with the HARPS spectrograph were performed under the ESO programme ID 082.C-0120, and observations with the VLT/UVES under ID 081.C-0413(C).

Author Contributions H.J.D. coordinated the analysis and its interpretation. P. Barge., S.A., J.M.A., R.A., J.C., L.C., T.M., M.O., M.P. and B. Samuel contributed to the treatment of the light curve and the detection of the transits in the CoRoT data. F.B., D.Q., C.M., G.H., M.M., C.L., F. Pepe, A.H., W.B., S.A., S.U. and F. Pont prepared, performed and analysed radial velocity observations; A. Baglin, M.A., J.S., L.J., P. Borde´. A. Le´ger, A. Llebaria and P.B. contributed fundamentally to the definition, design and operation of the CoRoT instrument. A.E., B.T., S.C., R.D., M.F., M.G., M. Hidas, T.L., H.R., D.R., R.S., A.S., H.J.D., R.d.H., R.A., M.R., P.K., B. Stecklum and D.C. performed ground-based photometry; Sz.C., R.A., M.B. and A. S. Bonomo worked on light curve modelling and parameter fitting. M.D., H.B., D.G., J.-C.G., E.G. and M.F. constitute the team that performed the stellar typing and related

observations. T.G., M. Havel, J.S., H.L., G.W. and S.F.-M. performed the modelling of the planet and the interpretation of its characteristics. All authors discussed the results and commented on the manuscript.

Author Information Reprints and permissions information is available at www.nature.com/reprints. The authors declare no competing financial interests. Correspondence and requests for materials should be addressed to H.J.D. (hdeeg@iac.es).

NATURE|Vol 464|18 March 2010 LETTERS

387

Macmillan Publishers Limited. All rights reserved